The invention relates to a control method with a characteristic line defined by interpolation points.
Processes overridden by strong, time-variant disturbances, in general, can be only insufficiently conducted by adding a linear regulator. To obtain the required excellence, there have to be used, as a rule, non-linear control loop structures and includes a pattern for disturbances. This often results in unstable situations. In practice, the disturbance pattern is mostly unknown and a finding is possible only at considerably expense, but often is also absolutely impossible.
In such cases, reference is had to a controlled standard of correcting variables on the basis of a stationary, empirical process pattern with a characteristic curve or a characteristic field without first taking into account the disturbances. The resulting guiding behavior is stationarily accurate for an undisturbed operation case and in the dynamic transition, very quick and yet stable. In case, disturbances occur stationarily unforeseeable deviations of s a greater or smaller magnitude, generate between the regulating variable and the command variable. Therefore, it has to be ensured, e.g. by limiting the correcting variables, that no critical operating states occur. Of course, it is no longer possible, in this case, sufficiently to meet the general requirements relative to stationary precision.
The invention is based on the problem of stably and with sufficient accuracy controlling a process that are overridden by strong, time-variant disturbances.
According to the invention, a control method is selected having a characteristic curve defined by interpolation points. The interpolation points can be empirical values obtained by testing, however they can also be arbitrarily preset, since they are adapted during the operation specifically in accordance with the system deviation of a correcting variable and a ratio of the difference between a measured point and an interpolation point and the difference between the two interpolation points adjacent to the measured point.
Correction values needed for the adjacent interpolation points are conveniently figured out from the following equations:
Ku=Rabw*[1xe2x88x92(xxe2x88x92xu)/(xoxe2x88x92xu)
and
Ko=Rabw*[1xe2x88x92(xoxe2x88x92x)/(xoxe2x88x92xu)
wherein Ko, Ku are correction values respectively for the lower and the upper interpolation points, x is an actual measured operating point on the characteristic curve and xo, xu values of a closest upper or lower interpolation point based on the measured value. The new adapted characteristic curve results from the initial interpolation points plus the appertaining correction values.
To increase the stability of the method and lightly to load the processing unit, it is convenient to perform the adaptation with a scanning time of more than 20 milliseconds. The scanning time must, if need be, adapted to the prevailing method time constant. It is further advantageous to limit the adaptation range of the characteristic curve in which an appertaining sensor works linearly. This is simpler in a control than in a regulation.
According to another embodiment of the invention, the new interpolation points are preset in steps. Excessively great bounces are thereby prevented during the operation and the output variable of the control proceeds less hectically.
From the correction values conclusions can be drawn on the operating behavior. Thus, correction values can be used for diagnosis purposes when, e.g. they exceed the preset limit values. On the other hand, defined theoretical values can be preset in order to test the reaction of the systems.
The inventive control method is especially adequate for pressure control of an automatic transmission, in particular, one with continuously variable ratio. The hydraulic adjustment of the static friction force between two friction elements such as an alternate belt and bevel pulleys is overridden by numerous time-variant disturbances. It must be dynamically and stationarily ensured concerning the contact pressure of the bevel pulleys that the belt, such as a sliding link belt, in no case skids over the bevel pulleys. The bevel pulleys and mainly the belt would thereby be destroyed. In addition, the contact pressure force produced by a hydraulic pressure decisively influences the efficiency of the transmission and thus directly the fuel consumption of the driven motor vehicle. Therefore, in the control of the contact pressure a reinforced crucial point is that there be kept the stationary accuracy. The inventive control method satisfies the requirements in relation both to the stability of the control and to the accuracy.
When the correction value gives reason to conclude that an error exists in the system, e.g. when it exceeds a preset limit value, it is convenient to activate a substitute function of the transmission, such as an emergency program. When the current supply is switched off, e.g. when the ignition of an internal combustion engine is switched off or the current supply fails, it is further advantageous to store the data in a memory module (e.g. EEPROM, FLASH) in order that they are again available when restarting. During the new start, if the last adapted characteristic curve is used, a long-time adaptation of the characteristic curve is obtained during the service life.
The inventive control method can be used not only with characteristic curve but in the same manner with n-dimensional characteristic fields. The correction values of the interpolation points are separately figured out for each dimension.